Thanks for the responses. I have to admit that I am still fresh-newbie to SA administration, however, the "integration" method is simply to pipe mail from citadel to spamd, which happens to be on the same server, and then if the filtration passes, then the mail gets passed back to the email server.
I am using Citadel as my email server, so the integration is simply to ad the a relavant host declaration, and that is about it. I think the process is as simple as described, however, I have never seen any spamd/SA headers in anything that gets passed back to the server. I can see the email getting filtered in the spamd logs, and when they were on the maillog facility, as indicated earlier, you could see the direct flow of an email coming into citadel, a connection being registered from citadel to spamd/SA, and then the processing of the email is displayed. I have provided an example directly off of the server logs. tail /var/log/spamd.log Fri Apr 10 22:39:33 2009 [26125] info: spamd: connection from localhost [127.0.0.1] at port 58085 Fri Apr 10 22:39:33 2009 [26125] info: spamd: checking message <01c9ba89 $6ab06200$1f3c7...@ter> for (unknown):1004 Fri Apr 10 22:39:37 2009 [26125] info: spamd: identified spam (22.3/5.0) for (unknown):1004 in 4.7 seconds, 1184 bytes. Fri Apr 10 22:39:37 2009 [26125] info: spamd: result: Y 22 - BODY_ENHANCEMENT2,FH_HELO_EQ_D_D_D_D,FM_SEX_HELODDDD,HELO_DYNAMIC_IPADDR,HS_INDEX_PARAM,MORE_SEX,MSGID_FROM_MTA_HEADER,RAZOR2_CF_RANGE_51_100,RAZOR2_CF_RANGE_E8_51_100,RAZOR2_CHECK,RCVD_IN_PBL,RCVD_IN_SORBS_DUL,RDNS_DYNAMIC,URIBL_BLACK,URIBL_JP_SURBL,URIBL_SBL scantime=4.7,size=1184,user=(unknown),uid=1004,required_score=5.0,rhost=localhost,raddr=127.0.0.1,rport=58085,mid=<01c9ba89$6ab06200$1f3c7...@ter>,autolearn=spam Fri Apr 10 22:39:38 2009 [26123] info: prefork: child states: II Fri Apr 10 22:56:41 2009 [26125] info: spamd: connection from localhost [127.0.0.1] at port 59165 Fri Apr 10 22:56:41 2009 [26125] info: spamd: checking message <01c9ba9c $92f58c00$bba3c...@tenaude> for (unknown):1004 Fri Apr 10 22:56:43 2009 [26125] info: spamd: identified spam (7.2/5.0) for (unknown):1004 in 2.2 seconds, 1561 bytes. Fri Apr 10 22:56:43 2009 [26125] info: spamd: result: Y 7 - MSGID_FROM_MTA_HEADER,RCVD_IN_BL_SPAMCOP_NET,RCVD_IN_XBL,SPF_SOFTFAIL scantime=2.2,size=1561,user=(unknown),uid=1004,required_score=5.0,rhost=localhost,raddr=127.0.0.1,rport=59165,mid=<01c9ba9c$92f58c00$bba3c...@tenaude>,autolearn=no Fri Apr 10 22:56:43 2009 [26123] info: prefork: child states: II <Prompt> cat /var/log/maillog | grep tenaude Apr 10 22:56:43 dataserver1 citadel: -1: from=<tena...@ssss.gouv.qc.ca>, nrcpts=1, relay=119.207.163.187 [119.207.163.187], stat=550 message rejected by spam filter^M I have reinstalled SpamAssassin using the perl modules that were listed as optional, and it seems that the sa-update worked, and I have not gotten anymore spam since reinstalling, then upgrading. I am still a bit vague as to how all of the integration and rulesets work, however, I think that the integration of SA with citadel is simply to pipe unknown email to the filter, and then let the filter manage it. There seems to be a notification within the citadel logs which indicates a -1 status and the requisite "rejected by filter" message, however, I don't see where the mail is held. Hence my interest in getting further involved in the administrative tasks, since time is permitting, now. Thanks for the inquiries. I hope that my current message will clear up the "integration" inquiry. I still have to read more about these different mechanisms, however, I think it may be working now. However, I do want to know about the "tagging" facilities that I am supposed to be seeing. Any further information on that topic is much appreciated, because spamd is obviously doing its job because int the time that it took to take spamd down, remove it, recompile, and install it, I recieved about fifteen spam mail messages. So there must have been something which was out of date on the system, because I can see about five to ten messages every 15 to 30 minutes being deflected in the logs. Thanks guys, and please give any further input or inquiry, because I am all ears.... On Fri, 2009-04-10 at 20:49 -0400, Karsten Bräckelmann wrote: > On Fri, 2009-04-10 at 19:24 -0400, martes wrote: > > > There was a mention of evolution's junk plugin, however, I had to > > disable that plugin and just rely on the server, since it would just > > cause an infinite loop, whenever new mail was looked at, causing > > Evloution to lock up. > > While the "loop" is disturbing and can't be caused by either system > involved -- I just asked, cause your headers are lacking the SA ones. A > possible explanation. Though the *least* best one, since you're using > IMAP. > > Yes, server side filtering *before* the mail gets delivered to your IMAP > server is the way to go. > > > > I use spamd on my mail server, and the server pipes it straight to my > > spamd session, and then if it gets out, then I get email. That may be > > the reason for the non-sa headers. I just looked at a few other "good" > > email examples in my inbox, and non of them has SA headers, except for > > the ones from this list. > > Nope. Using spamc/d server-side is *not* the reason why you're lacking > the SA headers. > > Let me ask again: HOW do you integrate SA? > > SA headers appearing on this list are normal, and have not been added on > your side. Besides, your ones would be at the top and not labeled ASF... > >
